j-can-see
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| J_SEE_MODEL | Yes | Vision model name (must be set explicitly) | |
| J_SEE_TOKEN | Yes | Vision model API key (not hardcoded — must be set explicitly) | |
| J_SEE_API_SPEC | No | Upstream API spec: responses / openai / anthropic | responses |
| J_SEE_BASE_URL | Yes | Vision endpoint base URL (must match J_SEE_API_SPEC; trailing slashes are stripped) | |
| J_SEE_MAX_EDGE | No | Max long-edge pixels for image compression | 1568 |
| J_SEE_MAX_BYTES | No | Max source file size in bytes; larger is rejected | 52428800 |
| J_SEE_REASONING | No | Reasoning effort (only honored by the openai spec): none / low / medium / high | none |
| J_SEE_TIMEOUT_MS | No | Vision call timeout in milliseconds | 90000 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| see_imageA | 读取图片（本地文件 / URL / 剪贴板 / 最近截图）并通过视觉模型返回文字描述。用于主模型无多模态输入能力时的识图。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 1 tool
With only a single tool, there is no possibility of confusing it with any other tool. The tool's purpose is clearly defined around image understanding.
The tool follows a clear verb_noun pattern (see_image), and since it is the only tool, the naming is internally consistent.
A single tool feels thin for a server, but the scope is narrow (image description). It is borderline but not an extreme mismatch.
The tool covers all input methods mentioned (local file, URL, clipboard, recent screenshot) and fulfills the stated purpose of describing images for models without multimodal input.